An interesting experiment with JSON, Postgres and Omnis
omnis at lineal.co.uk
Mon Jan 16 02:44:27 EST 2017
An age old problem then.
Currently you need just a few bits and pieces available in JS, just as we do. But what is the line? And how long long will that line last?
I remember going from 07 single library to O$, multi library and then SQL from DF1, all at the same time! No point in not, and we took 3 goes at doing it, throwing away work that was started.
I suspect that in the long run, 5 years maybe, we will want to have just one development platform, and all of the software running in that platform. But we will take small steps this time, just like yourselves.
One thing for sure, although Omnis has evolved over 30 years, I have stayed with it. I can say I haven't changed from Omnis, but Omnis sure has, and has carried my code with it. Mostly. But if I were to change over to a new set of tools, how transportable will my code be over the next 15 years? If you go back 15 years, which of those languages and tool sets are still supported? How convertible is the work done in those environments to today current best set? And how good will the change be in another 5, 10 and 15 years be? Seamless? Not that the new tools won't be bad, it is just the work involved in the migration. A younger man's job maybe.
My last point about swapping tools regularly relates to our product SQLWorks. It is a vertical solution that we sell. It does a lot, but it is a single tool. If your business is to write new solutions for different clients, then you can change tools often I would imagine, and transporting bits of precious code with you, updating where necessary. You will come across different kinds of problems that require different solutions.
But it is never an easy choice for sure.
> On 16 Jan 2017, at 00:26, Paul Mulroney <pmulroney at logicaldevelopments.com.au> wrote:
>> On 14 Jan 2017, at 1:00 AM, omnisdev-en-request at lists.omnis-dev.com wrote:
>> My point is merely that I've talked to plenty of developers whose needs
>> are small enough that learning a completely new environment is complete
>> overkill compared to creating a couple of remote forms in a technology
>> they already posses the skills in. Not everyone is in the market for
>> building a full fledge all bells and whistles website and mobile behemoth.
> We're in that boat - we have a thick client application, that only requires certain features to be web based/mobile based, so it's relatively painless to go from Omnis thick client to wmClient or jsClient.
> We're doing a slightly bigger project, and for that I'm looking at using Bootstrap for the website part, Omnis as the middleware and Postgres as the backend. We're planning to build a RESTful API in Omnis and have the website use that. We'll still have thick clients connected to the same system as well, so it's going to be an interesting one!
> Broken pencils are pointless.
> Paul W. Mulroney We Don't Do Simple Pty Ltd
> pmulroney at logicaldevelopments.com.au Trading as Logical Developments
> www.logicaldevelopments.com.au ACN 161 009 374
> Ph: +61 8 9458 3889 86 Coolgardie Street
> BENTLEY WA 6102
> Manage your list subscriptions at http://lists.omnis-dev.com
More information about the omnisdev-en